check_integration

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnly=false, destructive=false, idempotent=true, openWorld=true; the description adds far more: the two-tier model (pk_ vs sk_dev), that sk_live is refused by design, that the only writes are Tier B probe records in dev which are deleted at the end, and that dev/live enforce identical rules. It never states an explicit required permission scope or the exact runtime/rate profile beyond the timeout param, but the disclosure is well above the annotation baseline.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the actionable timing guidance, and each sentence carries real content (tiers, safety guarantees, failure meaning). It is dense and slightly repetitive in reasserting the sk_live prohibition, but not padded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 7-parameter tool with no output schema, the description covers what matters: which keys trigger which tier, what writes occur and that they are cleaned up, the dev/live equivalence argument, and the return shape. An agent has enough to call it correctly and interpret results.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description goes further by tying parameters to behavior — public pk_ enables Tier A, adding sk_dev enables Tier B, publicCollection confirms anonymous readability, and testUsers are the two Tier-B sessions. It does not add syntax/format detail beyond the schema, but the tier mapping is genuinely useful semantic context.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

Names a specific action (runs reaffirm boundary checks live against the caller's own app) and a concrete return shape (structured pass/fail with checks, notes, failedCount, passed). It is clearly distinguishable from the doc-oriented siblings (guide, reference, explain_rule, validate_collection_name) since it executes live checks rather than explaining concepts.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit timing windows: after wiring the app and before declaring it done, and again before go-live. It also states the remediation condition ('a failed check means the app's assumptions drifted — fix before shipping'), so the agent knows exactly when to invoke and what to do with the result.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.